Project Abstract Early in life many children develop wheezing which can be a sign of early-onset asthma. Yet, not all children who experience wheezing episodes develop asthma. Predicting asthma that begins early in life is important as those who develop early-onset asthma are more likely to have persistent symptoms and structural changes to the lung. Identification and early intervention of high risk individuals could prevent long term lung function abnormalities and provide an alternative to costly treatment. To date, risk prediction models for early- onset asthma are limited mainly to clinical history and presentation. Additionally, no one model contains both reliable sensitivity and specificity or clinically valid serological predictors for early-onset asthma. Our laboratory has focused previously on immunologic responses to common asthmatic triggers, house dust mite (HMD) and cockroach allergens, between individuals and how these responses differ between populations and outcomes. We have identified increased lymphoproliferation and cytokine response (IL-13) in allergen-stimulated cord blood mononuclear cells (CBMC) derived from individuals who are atopic or African- Americans. Bacterial products, such as lipopolysaccharide and peptidoglycan, stimulate the cytokine production of IL-13 and IFN-?. In this proposal our preliminary data indicate that the production of IL-13 after allergen exposure in CBMCs is associated with exposure to different microbiota. In addition, early exposure to Moraxellaceae greatly enhances IL-13 production by CBMCs after exposure to HDM. Taken together, our data suggest that early-life bacterial exposure may enhance the production of cytokines that can alter host physiology and immunity. In this translational medicine proposal, we will build a predictive model for early-onset asthma that will be applicable to general and high risk populations. Our approach will include high-throughput techniques to assess bacterial exposure and immunologic response in the perinatal time period. We will integrate the large amount of data generated into a pipeline with machine learning algorithms that will be regulated specifically for the prediction of later onset of disease. In addition, we will assess bacterial exposure in association with mothers who carry risks for asthma development in their children. The proposed studies will address both the need of predictive model for early-onset asthma and identify bacterial-host interactions associated with high risk populations (e.g. maternal risk factors). Together, these studies form the platform for multidisciplinary training in microbe- immune interactions, predictive risk for asthma, bioinformatics and form a solid foundation to launch a successful career as a physician- scientist.